Performance Monitoring, Assessment and Modification of an Adaptive MPC: Automated Insulin Delivery in Diabetes

In this work, a controller performance monitoring, assessment, and modification (CPMAM) system is developed to analyze the closed-loop behavior and modify the parameters of a model predictive control (MPC) system. To this end, various performance indices are defined to quantitatively evaluate the controller efficacy in real-time. The CPMAM module also incorporates online learning from historical operational data to anticipate impending disturbances and proactively counteract their effects. The MPC is used in automated insulin delivery systems for people with Type 1 diabetes, called artificial pancreas systems.